APEX
Leaderboard
Models
Compare
Tasks
Metrics
About
Leaderboard
Overall
Frontend
Backend
Full-Stack
Debugging
Refactoring
Code Review
From Scratch
Multi-Language
All Levels
Easy
Medium
Hard
Expert
Master
#
Model
ELO
Peak
Avg Score
Avg Cost
Consistency
1
Claude Opus 4.8
1905
2057
90.0
$2.04
100.0%
2
Claude Opus 5
1865
2442
89.2
$0.61
100.0%
3
Claude Opus 4.7
1839
1984
88.3
$1.47
100.0%
4
GPT 5.5
1801
1952
86.8
$1.61
94.3%
5
Deepseek V4 Flash 0731
1763
2056
85.7
$0.04
100.0%
6
GLM 5.2
1755
1911
85.3
$0.39
97.1%
7
Claude Sonnet 5
1731
2026
84.7
$0.41
98.6%
8
GPT 5.4 Mini
1729
1876
84.3
$0.38
95.7%
9
Composer 2.5
1727
1877
84.7
$0.39
100.0%
10
Claude Opus 4.6
1718
1871
84.7
$1.11
97.1%
11
GPT 5.2
1717
1717
80.2
$0.19
84.4%
12
Kimi K2.7 Code
1708
1864
83.9
$0.60
97.1%
13
Claude Sonnet 4.6
1703
1848
83.8
$0.31
91.4%
14
Grok 4.3
1702
1854
83.5
$0.39
97.1%
15
Minimax M3
1701
1840
83.9
$0.15
98.6%
16
GPT 5.3 Codex
1693
1695
80.2
$0.13
82.8%
17
Claude Opus 4.5
1660
1815
82.3
$0.88
92.9%
18
GPT 5.2 Codex
1656
1694
78.5
$0.13
80.3%
19
Kimi K2.6
1639
1793
80.8
$0.47
91.4%
20
GLM 5.1
1630
1778
80.7
$0.22
92.9%
21
Qwen3.7 Max
1627
1771
80.2
$1.00
92.7%
22
Deepseek V4 Pro
1614
1771
80.2
$0.37
90.0%
23
GPT 5.3 Codex Spark
1599
1755
79.3
$0.32
92.9%
24
Deepseek V4 Flash
1598
1745
79.6
$0.06
88.6%
25
Gemini 3.5 Flash
1596
1748
79.8
$0.27
94.3%
26
GPT 5.1 Codex Mini
1587
1706
76.6
$0.59
82.2%
27
Minimax M2.7 [NVFP4]
1578
1733
79.4
<$0.01
93.5%
28
Qwen3.6 27b [Q4_K_XL]
1571
1724
78.8
<$0.01
87.1%
29
Gemini 3.1 Pro Preview
1570
1682
75.9
$0.57
70.3%
30
Qwen3.6 Plus
1564
1627
77.8
$0.40
90.0%
31
Gemini 3 Pro Preview
1549
1647
74.9
$0.46
72.3%
32
GLM 5
1539
1650
73.2
$0.15
69.4%
33
Qwen3.6 35b A3b Q4 K XL [Q4_K_XL]
1531
1677
76.5
<$0.01
85.9%
34
Kimi K2.5
1515
1540
72.4
$0.06
68.9%
35
Claude Sonnet 4.5
1513
1678
76.1
$0.25
72.3%
36
Grok 4
1511
1526
71.8
$0.27
66.9%
37
Minimax M2.7
1507
1673
76.4
$0.05
78.6%
38
Qwen3.5 27b
1503
1566
70.0
$0.36
59.8%
39
GLM 4.7
1499
1604
71.8
$0.10
64.5%
40
Qwen3.6 35b A3b [BF16]
1499
1662
76.2
<$0.01
77.1%
41
Qwen3.5 Plus 02.15
1486
1593
70.6
$0.13
56.5%
42
Qwen3.5 122b A10b
1485
1559
69.6
$0.38
57.3%
43
Qwen3.5 122b A10b [Q4_K_XL]
1484
1618
71.8
$0.24
60.2%
44
Gemini 2.5 Pro
1484
1532
68.4
$0.27
53.4%
45
Qwen3.6 27b [BF16]
1481
1640
73.2
<$0.01
67.5%
46
Qwen3.5 397b A17b Q4 K XL [Q4_K_XL]
1476
1636
74.8
$0.85
75.7%
47
Grok 4.1 Fast
1473
1473
68.7
$0.05
60.2%
48
Claude Haiku 4.5
1455
1621
71.5
$0.07
63.1%
49
Gemini 3 Flash Preview
1436
1609
72.4
$0.02
67.4%
50
Minimax M2.1
1416
1441
65.1
$0.05
46.4%
51
GLM 4.7 [Q4_K_XL]
1399
1589
71.2
$0.04
56.1%
52
Minimax M2.5
1394
1422
65.5
$0.14
45.5%
53
Step 3.5 Flash
1386
1438
60.4
$0.13
39.8%
54
Qwen3.5 Flash 02.23
1382
1485
63.4
$0.06
46.3%
55
Grok Code Fast 1
1381
1409
65.8
$0.07
42.4%
56
Qwen3.5 27b [Q4_K_M]
1369
1467
63.1
$0.18
42.7%
57
Qwen3.5 35b A3b
1367
1494
63.2
$0.08
42.1%
58
GLM 4.5
1365
1494
65.9
$0.10
49.5%
59
GLM 4.6
1364
1496
64.4
$0.11
40.7%
60
Deepseek R1 0528
1363
1505
64.6
$0.05
43.7%
61
Qwen3 Coder
1360
1388
60.8
$0.11
37.8%
62
Qwen3 Coder Plus
1360
1489
63.2
$0.07
39.3%
63
Qwen3 Coder Next
1318
1453
61.2
$0.02
36.1%
64
Deepseek V3.2
1317
1495
64.0
$0.04
37.5%
65
Devstral 2512
1314
1454
63.0
$0.10
32.4%
66
GPT OSS 120b
1290
1378
59.1
$0.12
30.2%
67
Minimax M2.5 [Q4_K_XL]
1290
1502
63.9
$0.03
33.3%
68
GLM 4.5 Air
1284
1430
60.3
$0.03
29.1%
69
Qwen3 Coder Flash
1278
1309
59.1
$0.02
27.5%
70
Trinity Large Preview:free
1266
1413
54.5
$0.05
20.3%
71
Qwen3.5 35b A3b [Q4_K_XL]
1222
1355
49.4
$0.05
17.2%
72
GLM 4.7 Flash
1217
1370
55.8
$0.01
22.4%
73
Nemotron 3 Nano 30b A3b
1201
1239
52.3
$0.09
18.3%
74
Gemini 2.5 Flash Lite
1201
1401
56.6
$0.02
21.5%
75
GPT OSS 20b
1197
1308
53.8
$0.11
21.9%
76
Qwen3 Coder Next [Q4_K_XL]
1192
1408
58.3
<$0.01
19.5%
77
Qwen3 Coder 30b [Q4_K_M]
1035
1312
52.1
$0.01
12.5%
ELO Distribution